a 
J. W. Mallet on the Atomic Weight of Lithium. 358 
with advantage, using a larger amount of a carefully prepared 
salt, better adapted to the purpose than the sulphate, and requir- 
ing the process of analysis to be varied. The salt chosen was 
the chlorid of lithium, and I shall deseribe first the mode of its 
preparation, and then the experiments which have been made 
upon its composition. 
Crystallized spodumene from the granite of Goshen (Mass.), 
where the mineral occurs with blue tourmaline, beryl, and rose 
mica, was finely pulverized, and 1 part mixed with 8 or 4 of 
unslaked lime and about three-fourths of sal-ammoniac. The 
mixture was heated in large crucibles to the highest temperature 
of a good wind-furnace. ‘This is the process proposed by Prof. 
J. L, Smith* for the analysis of silicates, except that he uses car- 
bonate of lime instead of the caustic earth. The nearly fused 
mass was pulverized, mixed with water, and treated with sul- 
was boiled down to a moderate bulk, and ag Se with chlo- 
nid of barium; the sulphate of baryta thrown down was was 
solved, the solution boiled with a little pure milk of lime, and 
filtered from magnesia. The lime in the filtrate was removed 
by oxalate of ammonia, and the solution was evaporated to dry- 
little water and again dried at a gentle heat, not over 100° C. 
The dry mass was A into a glass-stoppered bottle, and 
ured upon it—solution of the chlorid of lithium being aided 
shaking the bottle from time to time. After a few hours the 
Clear liquid was decanted, and the alcohol and ether were dis- 
ed : 
of purification was re eated a third time, nothing being now left 
Whdissolved by the siaweknoiels The pure chlorid of lithium 
* Amer, Jour. of Science, [2], xv, 284; xvi, 53. 
SECOND SERIES, VOL. XXII, NO. 66,—-NOV., 1856. 
45 
